利用案例式推理來輔助電腦問題診斷

Abstract

隨著時代的變遷，資訊科技在企業內已經扮演一個不可或缺的角色。企業內部應用資訊科技所建置的系統日趨複雜，因此一些使用電腦產生的問題也越來越多，所投入的人力和解決問題的時間對於企業來說都是成本。在面對資訊人員的流動率過高及專業知識分散不易整合的問題上，如何能將使用者的問題描述作一整理和正確的歸類，以便能找出最合適或相近的處理案例來解決問題，是重要的研究議題。 案例式推理主要是運用先前的案例，作為解決日後問題的參考的依據。案例式推理應用的範圍相當廣泛，包括客服系統(Help Desk)的應用。若能結合文字探勘的技術，將有效率地從大量文字性資料中整理出有用的資訊。本研究是利用文字探勘(Text Mining)中的文件自動分類技術來取代人工分類。再結合案例式推理方法(CBR)來尋找最相近的案例來解決資訊系統所遇到的問題。本研究最後建構一個雛型系統針對所提方法做一比較驗證。藉此期望對企業內部客服系統(Help Desk)解決個人電腦問題能有所提升，並強化公司整體競爭力。

Enterprises put emphasis on Information Technology to support business operations. The more complex internal IT application used by Enterprise leads to more computer operation problems. For Enterprises, human resource and time for trouble shooting are costs. It is an important issue for enterprises to correctly categorize problems and find out the most suitable cases for trouble shooting, when they facing the problems of high replace rate of IT members and too scattered know how. Case-Based Reasoning is the process of solving new problems based on the solutions of similar past problems. CBR is widely used in help desk system. This work integrates CBR with Text Mining technology to extract useful information from text-based documents. We use Text Mining technique to categorize problems, and then we use CBR to look for the most similar cases in selected category to solve the new problem. A prototype system is implemented to demonstrate the effectiveness of our approach. This work will support Help Desk System to solve problem efficiently and quickly, and it will also strengthen the enterprise competitiveness.